Retrieval-Augmented Generation for LLMs

Business Features

• F1: Give the LLM external tool access
• F2: Customize the LLM with local data
• F3: Implement chat based memory - #4

Experimentation

• following
• "Augumenting Your LLM using Retrieval Augumented Generation" https://learn.nvidia.com/courses/course-detail?course_id=course-v1:DLI+S-FX-16+V1
• https://python.langchain.com/v0.2/docs/tutorials/
• https://python.langchain.com/v0.2/docs/tutorials/rag/
• https://python.langchain.com/v0.2/docs/tutorials/local_rag/

Documentation

• ObrienlabsDev/blog#47
• 2020 paper where RAG as a term was introduced by Patrick Lewis et al. https://arxiv.org/abs/2005.11401
• 2017 paper outlining retrieval/generation via Wikipedia from Chen et al. https://arxiv.org/abs/1704.00051
• 2020 paper on RAG from Patrick Lewis - https://arxiv.org/pdf/2005.11401
• 2024 paper on "Retrieval-Augmented Generation for Large Language Models: A Survey" https://arxiv.org/pdf/2312.10997
• google search "rag ai wikipedia"
• https://blogs.nvidia.com/blog/what-is-retrieval-augmented-generation/
• https://en.wikipedia.org/wiki/Retrieval-augmented_generation
• https://towardsdatascience.com/what-nobody-tells-you-about-rags-b35f017e1570
• https://alexgarcia.xyz/blog/2024/sqlite-vec-stable-release/index.html
• book: AI Engineering by Chip Huyen Jan 2025

Alternatives to RAG

• Long term memory - revisting memorization - #10